Country
ASPEN: Breaking Operator Barriers for Efficient Parallel Execution of Deep Neural Networks
ASPEN also achieves high resource utilization and memory reuse by letting each resource asynchronously traverse depthwise in the DNN graph to its full computing potential.
Diffusion-Reward Adversarial Imitation Learning
Despite its encouraging results, GAIL training is often brittle and unstable.
On the Use of Anchoring for Training Vision Models
The open-source code is available at - https://software.llnl.gov/anchoring
ProG: A Graph Prompt Learning Benchmark
The first two authors contributed equally to this work.
Would I Lie To You? Inference Time Alignment of Language Models using Direct Preference Heads
Pre-trained Language Models (LMs) exhibit strong zero-shot and in-context learning capabilities; however, their behaviors are often difficult to control.